Find open-source science resources

Long-context generative genomic foundation model using 6-mer tokenization for DNA sequence modeling and generation, with v2 model families for prokaryote and eukaryote genomes and pretrained weights available on HuggingFace (GenerTeam, 460+ stars, MIT License, 2025-2026)

Democratizing AlphaFold3: PyTorch reimplementation to accelerate protein structure prediction research

Pretrained time series foundation model for zero-shot forecasting across diverse scientific and real-world domains; tokenizes continuous time series into discrete bins to train transformer language models on large-scale corpora, achieving strong zero-shot generalization and competitive performance with task-specific supervised models on climate, energy, and health benchmarks (5.3K+ stars, Apache 2.0, 2024-2026)

Medical time series foundation model pretrained on 454B time points from heterogeneous clinical corpora spanning ICU physiological signals and hospital EHR, with continuous-time rotary positional encoding, frequency-specialized Mixture-of-Experts, and neural ODE extrapolation for zero-shot forecasting across irregular and multimodal temporal health data (Microsoft, 399+ stars, MIT License)

MCP server, CLI, and agent skills for searching and downloading academic papers from multiple open sources (arXiv, PubMed, bioRxiv, Semantic Scholar, OpenAlex, CORE, Europe PMC, etc.) with unified, deduplicated, LLM-friendly retrieval and an OA-first download fallback chain (OpenAGS, 1.9K+ stars, MIT License, 2025)

Open software framework for Engineering AI built on transformer building blocks, enabling teams to build, train, and operate industrial simulation models across engineering verticals; includes ready-to-use recipes for CFD (AB-UPT on DrivAerML), external aerodynamics, and heat transfer (234+ stars, ENPL non-commercial license, 2026)

First agentic LLM for autonomous data science with end-to-end pipeline from data to analyst-grade reports

All-atom generative world model for all-to-all biomolecular interaction design, enabling cross-modality generation of proteins, nucleic acids, small molecules, and cyclic peptides with fine-grained epitope-level control and 2-4 orders of magnitude faster design throughput than modality-specific baselines (316+ stars, Apache 2.0)

Multi-modal foundation model for biomolecular structure prediction (proteins, small molecules, DNA, RNA, glycans) achieving SOTA across benchmarks, with optional MSA/template support (Chai Discovery, 2024)

Self-hostable scientific claim-verification and literature-review tool combining Semantic Scholar retrieval, bibliometric scoring, and LLM-based evidence synthesis for large-batch validation workflows

Deep learning with spiking neural networks in Python, providing gradient-based training of SNNs via PyTorch autodifferentiation for brain-inspired computing and neuromorphic research, with online learning capabilities and extensive tutorials (1.9K+ stars, actively maintained)

Learnable latent embeddings for joint behavioral and neural analysis, enabling consistent and interpretable mapping of neural activity to behavior across modalities, species, and experiments (EPFL & Harvard, 1K+ stars)

Vision foundation model for the tree of life, pretrained on diverse biological imagery across taxa for zero-shot species identification, trait extraction, and biodiversity research (Ohio State University Imageomics Institute)

Automated academic illustration generation for AI scientists, converting research papers into publication-ready figures using VLMs and diffusion models with iterative refinement (PKU & Google Research, 6.2K+ stars, 2026)

First multi-domain generative foundation model for the natural sciences built on a unified scientific grammar, encoding proteins, antibodies, small molecules, chemical reactions, materials, and their spatial interactions into a shared token vocabulary; enables unified generation, prediction, and design across domains under a purely autoregressive paradigm (134+ stars, Apache 2.0, 2026)

Deep learning library for solving PDEs

Cross-domain foundation model for continuum dynamics trained on 19 physical scenarios spanning 63 variables, featuring adaptive compute via stride modulation and patch jittering for long-run stability (Polymathic AI, 293+ stars, MIT License)

Controllable foundation model for general and specialized biomolecular structure prediction across proteins, nucleic acids, and complexes, featuring a public web server for interactive prediction workflows (IntelliGen AI, 223+ stars, Apache 2.0, 2025)

Autonomous multi-agent AI scientist that mirrors a PhD workflow: literature review → grounded hypothesis → real ML experiments → self-critique → write-up; features a deterministic harness with frozen-metric scoring, edit allowlists, and a verified registry to make reward-hacking and hallucination impossible, plus 108 unit tests runnable without API keys or GPUs (461+ stars, MIT License, 2026)

Foundation model for universal prompt-driven medical image segmentation extending SAM3 to clinical imaging, supporting 2D public benchmarks and 3D training/evaluation with text and box prompts; pretrained weights available on HuggingFace (189+ stars)

Numerical differential equation solving in JAX

Fudan University's cascade machine learning forecasting system for 15-day global weather prediction, employing a 3D Earth-specific transformer with hard-constraint techniques to achieve state-of-the-art accuracy against traditional NWP and AI baselines

Diffusion-based document OCR framework replacing autoregressive decoding with block-level parallel diffusion decoding, enabling high-accuracy text recognition in scientific PDFs (613+ stars, MIT License)

Cross-platform system optimizations for accelerating AlphaFold3 training with 1.73x speedup and 1.23x memory reduction

Minimalist, batteries-included repository for training video world models with diffusion-forcing, supporting long-horizon rollouts, 3D point-cloud generation, and model-predictive control with pretrained checkpoints (Simchowitz Lab, 700+ stars, MIT License, 2026)

Learning the language of protein-protein interactions

Genomic foundation model for metagenomic and genome annotation, featuring an 8k base-pair context and 500M parameters trained on 386B base pairs of eukaryotic DNA; provides expert models and a unified CLI for prokaryotic/eukaryotic coding-sequence annotation with strong performance on Genomic Benchmarks, Nucleotide Transformer tasks, and custom Gener tasks (GenerTeam, 314+ stars, MIT License)

Automatic atomic model building program for cryo-EM maps using deep learning, enabling rapid de novo protein structure determination from electron density with high accuracy (3DEM/EMBL, 169+ stars)

Generalist deep learning algorithm for cell and nucleus segmentation across diverse image types, with human-in-the-loop training (2.0) and one-click image restoration (3.0), 70K+ training objects (Nature Methods 2021/2022/2025)

Python library to train, interpret, and apply deep learning models to DNA sequences, providing a unified framework for regulatory genomics with support for CNN and transformer architectures, variant effect prediction, and attribution analysis (325+ stars)

General-purpose deep learning backbone for molecular modeling

Universal foundation model for grounded biomedical image interpretation, enabling comprehensive visual understanding, reasoning, and grounding across diverse biomedical imaging modalities with strong zero-shot generalization (55+ stars, Apache 2.0, 2025-2026)

Sparse identification of nonlinear dynamics

Closed-loop multi-agent system from hypothesis to verification across 12 scientific tasks, #1 on MLE-Bench (36.44%)

Programmatic data labeling and weak supervision

Multi-agent system with Parser-Planner-Painter architecture converting `paper.pdf` to editable `poster.pptx`, outperforms GPT-4o with 87% fewer tokens

Family of causal genomic foundation models trained on 1T tokens (~6T DNA base pairs) from the Carbon Pretraining Corpus, combining eukaryote genes, mRNA transcripts, and prokaryote genomes with a hybrid text/6-mer tokenizer; Carbon-3B matches or beats Evo2-7B on zero-shot DNA evaluations including sequence recovery, variant effect prediction, and perturbations (Apache 2.0, 201+ stars)

State-of-the-art RNA 3D folding model developed with Stanford Das Lab and Kaggle competition winners, featuring a 488M-parameter AF3-like architecture with MSA and template-based modeling, enabling structure-driven drug discovery and RNA therapeutics design (NVIDIA-Digital-Bio, Apache 2.0)

LLM-native molecular language that represents molecules as explicit graph-based code, enabling LLMs to operate and reason on chemistry directly with 5× lower token cost and ~76-80% accuracy on novel molecules vs ~20% for SMILES; supports small molecules, polymers, and Markush structures with lossless RDKit interconversion and Claude Code/Codex agent skills (AtomFlow, arXiv:2605.16480, 281+ stars, MIT License, 2026)

Foundation model for universal cell segmentation achieving state-of-the-art performance across bacteria, tissue, yeast, cell culture, and diverse imaging modalities (brightfield, fluorescence, phase), with pip-installable inference and Napari plugin (vanvalenlab/Caltech, bioRxiv 2024)

Apache 2.0 single-cell foundation model family scaling to 3B parameters, pretrained on 266M cell profiles including perturbation data and released with training, embedding, and downstream benchmarking workflows for disease-relevant single-cell tasks (2025)

End-to-end composable multi-agent framework for automating OpenFOAM-based CFD simulations from natural language prompts, managing meshing, case setup, execution, error correction, and post-processing; achieves 100% success rate on 110 FoamBench tasks with Claude Opus 4.6 through Architect-Input Writer-Runner-Reviewer agent collaboration with RAG-enhanced generation and MCP tool integration (RPI CSML, 242+ stars, MIT License)

Programmatic framework for designing state-switching proteins via backpropagation through compositional design constraints parameterized by structure prediction models; enables de novo design of allosteric regulators and fluorescent biosensors for arbitrary small-molecule analytes (79+ stars, MIT License, ICML 2026)

First fully open-source model achieving AlphaFold3-level accuracy with 1000x faster binding affinity prediction (MIT)

Highly scalable equivariant deep learning interatomic potentials enabling million-atom molecular dynamics simulations with ab initio accuracy, building on E(3)-equivariant architectures for large-scale atomistic modeling (mir-group, MIT License, 480+ stars)

Foundational auto-research agent framework for agentic science at scale, providing modular agent construction, run-level self-evolution, and multiple SciMaster domain agents (ML-Master, X-Master, Browse-Master); outperforms general-purpose agents across authoritative benchmarks including the OpenAI Frontier Science Benchmark (206+ stars, Apache 2.0, 2026)

Decentralized self-organizing teams of AI agents for long-running computational scientific experimentation; agents critique each other's proposals before spending compute and share successes/failures to avoid redundant exploration, achieving +8.33% on BioML-Bench, 1.9× faster nanoGPT optimization, and +12.5% on ProteinGym ACE2-Spike (425+ stars, 2026)

Generative foundation model for functional antibody and nanobody design, supporting de novo generation, affinity maturation, inverse design, structure prediction, and humanization (Tencent AI4S, ICLR 2025)

End-to-end autonomous AI research engine that turns an idea into a complete LaTeX paper by dispatching real computational experiments to local GPUs or SLURM clusters, collecting actual results, generating figures/tables, and writing a data-grounded manuscript rather than LLM hallucinations (OpenRaiser, 1.5K+ stars, MIT License, 2026)

Generalized biological foundation model with unified nucleic acid and protein language, integrating DNA/RNA/protein sequences (Nature Machine Intelligence 2025)